Let ViT Speak: Generative Language-Image Pre-training

Open paper
Benchmark Model Rank Results
optical-character-recognition-ocr-on-ocrbenchGenLIP-g/16 + LLaVA-NeXT-Qwen2.5-7B#64Score: 659
image-classification-on-imagenetGenLIP (g/16)#268Top 1 Accuracy: 85.2%

Rank counts only results with a code link.